FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2022122132> ?p ?o ?g. }

• W2022122132 endingPage "494" @default.
• W2022122132 startingPage "484" @default.
• W2022122132 abstract "Article1 February 1997free access The mammalian profilin isoforms display complementary affinities for PIP2 and proline-rich sequences Anja Lambrechts Anja Lambrechts Flanders Interuniversity Institute for Biotechnology, Department of Biochemistry, Faculty of Medicine, Universiteit Gent, 9000 Gent, Belgium Search for more papers by this author Jean-Luc Verschelde Jean-Luc Verschelde Flanders Interuniversity Institute for Biotechnology, Department of Biochemistry, Faculty of Medicine, Universiteit Gent, 9000 Gent, Belgium Search for more papers by this author Veronique Jonckheere Veronique Jonckheere Flanders Interuniversity Institute for Biotechnology, Department of Biochemistry, Faculty of Medicine, Universiteit Gent, 9000 Gent, Belgium Search for more papers by this author Mark Goethals Mark Goethals Flanders Interuniversity Institute for Biotechnology, Department of Biochemistry, Faculty of Medicine, Universiteit Gent, 9000 Gent, Belgium Search for more papers by this author Joël Vandekerckhove Joël Vandekerckhove Flanders Interuniversity Institute for Biotechnology, Department of Biochemistry, Faculty of Medicine, Universiteit Gent, 9000 Gent, Belgium Search for more papers by this author Christophe Ampe Christophe Ampe Flanders Interuniversity Institute for Biotechnology, Department of Biochemistry, Faculty of Medicine, Universiteit Gent, 9000 Gent, Belgium Search for more papers by this author Anja Lambrechts Anja Lambrechts Flanders Interuniversity Institute for Biotechnology, Department of Biochemistry, Faculty of Medicine, Universiteit Gent, 9000 Gent, Belgium Search for more papers by this author Jean-Luc Verschelde Jean-Luc Verschelde Flanders Interuniversity Institute for Biotechnology, Department of Biochemistry, Faculty of Medicine, Universiteit Gent, 9000 Gent, Belgium Search for more papers by this author Veronique Jonckheere Veronique Jonckheere Flanders Interuniversity Institute for Biotechnology, Department of Biochemistry, Faculty of Medicine, Universiteit Gent, 9000 Gent, Belgium Search for more papers by this author Mark Goethals Mark Goethals Flanders Interuniversity Institute for Biotechnology, Department of Biochemistry, Faculty of Medicine, Universiteit Gent, 9000 Gent, Belgium Search for more papers by this author Joël Vandekerckhove Joël Vandekerckhove Flanders Interuniversity Institute for Biotechnology, Department of Biochemistry, Faculty of Medicine, Universiteit Gent, 9000 Gent, Belgium Search for more papers by this author Christophe Ampe Christophe Ampe Flanders Interuniversity Institute for Biotechnology, Department of Biochemistry, Faculty of Medicine, Universiteit Gent, 9000 Gent, Belgium Search for more papers by this author Author Information Anja Lambrechts1, Jean-Luc Verschelde1, Veronique Jonckheere1, Mark Goethals1, Joël Vandekerckhove1 and Christophe Ampe1 1Flanders Interuniversity Institute for Biotechnology, Department of Biochemistry, Faculty of Medicine, Universiteit Gent, 9000 Gent, Belgium The EMBO Journal (1997)16:484-494https://doi.org/10.1093/emboj/16.3.484 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info We present a study on the binding properties of the bovine profilin isoforms to both phosphatidylinositol 4,5-bisphosphate (PIP2) and proline-rich peptides derived from vasodilator-stimulated phosphoprotein (VASP) and cyclase-associated protein (CAP). Using microfiltration, we show that compared with profilin II, profilin I has a higher affinity for PIP2. On the other hand, fluorescence spectroscopy reveals that proline-rich peptides bind better to profilin II. At micromolar concentrations, profilin II dimerizes upon binding to proline-rich peptides. Circular dichroism measurements of profilin II reveal a significant conformational change in this protein upon binding of the peptide. We show further that PIP2 effectively competes for binding of profilin I to poly-L-proline, since this isoform, but not profilin II, can be eluted from a poly-L-proline column with PIP2. Using affinity chromatography on either profilin isoform, we identified profilin II as the preferred ligand for VASP in bovine brain extracts. The complementary affinities of the profilin isoforms for PIP2 and the proline-rich peptides offer the cell an opportunity to direct actin assembly at different subcellular localizations through the same or different signal transduction pathways. Introduction Profilin is a ubiquitous, small (12–15 kDa) actin-binding protein that plays an important role in the regulation of actin polymerization in a number of motility functions (Carlsson et al., 1977; Stossel, 1993). In vitro experiments suggest that profilin can promote actin assembly from the G-actin–thymosin-4 pool when barbed filament ends are free. In the presence of gelsolin, a barbed end capping protein, profilin can act as an actin-sequestering protein resulting in actin depolymerization (Pantaloni and Carlier, 1993). This dual function of profilin is thought to be essential for rapid filament turnover during cell motility processes, e.g. in the growing end of fast moving lamellipodia of fibroblasts (Buss et al., 1992; Small, 1995) and at the tail of Listeria monocytogenes, a bacterial pathogen that uses actin assembly to move forward (Theriot et al., 1994). Profilin also binds phosphatidylinositol 4,5-bisphosphate (PIP2) (Lassing and Lindberg, 1985, 1988). Phospholipase C1 cannot hydrolyse profilin-bound PIP2 in vitro, unless the phospholipase is phosphorylated (Goldschmidt-Clermont et al., 1990, 1991; Machesky et al., 1990). These experiments led to a model in which profilin plays an important role as mediator in cell signalling to the cytoskeleton. PIP2-associated profilin is not able to bind actin monomers but, after hydrolysis of PIP2, profilin is free and can diffuse into the cytoplasm to interact with actin (Lassing and Lindberg, 1985, 1988). Profilins, with the exception of vaccinia virus profilin (Machesky et al., 1994), bind poly-L-proline, but the significance of this interaction remained an enigma for a long time. However, several proteins associated with the cytoskeleton or its dynamics contain such a sequence: cyclase-associated protein (CAP), a surface protein from the cytotoxic pathogen L.monocytogenes (ActA) and the vasodilator-stimulated phosphoprotein (VASP). Already in 1991, Vojtek et al. proposed a functional link between profilin and CAP, which is the 70 kDa subunit of the Saccharomyces cerevisiae adenylyl cyclase complex (Field et al., 1990). CAP is a bifunctional protein of which the N-terminal domain is necessary and sufficient for a Ras-responsive adenylyl cyclase complex, and of which the C-terminal domain binds actin (Freeman et al., 1995) and is important for normal cell morphology and responsiveness to nutrient extremes (Gerst et al., 1991). CAP mutants that lack the C-terminal domain can be rescued by overexpression of profilin (Vojtek et al., 1991). Since CAP has a proline-rich stretch in its middle domain, it may be a ligand for profilin (Goldschmidt-Clermont and Janmey, 1991). Similarly, ActA contains three single stretches of four proline residues. A Listeria mutant which does not express ActA on the surface is unable to form actin tails, to spread intracellularly and to infect adjacent cells (Domann et al., 1992), although host factors also seem to be required for actin tail formation (Theriot et al., 1994; Chakraborty et al., 1995). VASP was identified originally as a protein present in human platelets which becomes phosphorylated by cAMP- and cGMP-dependent protein kinases in response to both cAMP- and cGMP-elevating vasodilators and platelet inhibitors (Waldmann et al., 1986, 1987; Halbrügge et al., 1990; Nolte et al., 1991). The sequence reveals a central domain with a proline-rich motif, with five consecutive proline residues, that occurs as a single copy and a 3-fold repeat (Haffner et al., 1995). Only very recently, Reinhard and co-workers (1995) showed that VASP interacts directly with profilin. It was shown that human cells have two profilin isoforms and, given the different expression levels of each isoform in different tissues (Honoré et al., 1993), researchers have used, unknowingly, preparations containing different ratios of profilin isoforms in the past, e.g. in microinjection studies (Cao et al., 1992) and bio- and immunochemical studies (Buss and Jockusch, 1989). Human and bovine profilin I (Ampe et al., 1988; Kwiatkowski and Bruns, 1988) and II (Honoré et al., 1993; Lambrechts et al., 1995) show only 65.5% identity. Most of the mutated residues map at the exterior of the known profilin I structure, suggesting that both isoforms have a similar overall fold but display different biochemical properties. Their affinity for actin is quite similar (Lambrechts et al., 1995), consistent with the observation that the residues of profilin, forming the interface with actin, are conserved. However, one of our previous experiments suggested that the isoforms display a different affinity for poly-L-proline at acidic pH (Lambrechts et al., 1995). To probe this difference further, we investigated the interaction of each isoform with proline-rich model peptides derived from VASP, CAP and ActA using fluorescence spectroscopy at more physiological pH. We also searched for differences in interaction with the other known profilin ligand, PIP2, using a microfiltration assay. We show that profilin I has higher affinity for PIP2, while profilin II has higher affinity for proline-rich sequences, and can form dimers upon binding these sequences. In addition we show that PIP2 is an effective competitor for poly-L-proline binding of profilin I, but not of profilin II. Our data suggest that in cells profilin I may be preferentially associated with PIP2, and profilin II with proteins such as VASP, which we show is recruited preferentially from bovine brain extracts by profilin II. Results Bovine profilin I has a higher affinity for PIP2 than the profilin II isoform An important characteristic of profilins is their ability to bind PIP2 (Lassing and Lindberg, 1985, 1988). Our previous results showed that bovine profilin II binds PIP2, but nevertheless they hinted at a reduced affinity compared with profilin I (Lambrechts et al., 1995). To investigate this potential difference between the profilin isoforms further, we used an assay described by Haarer et al. (1993). In a microfiltration experiment, we incubated a constant amount of profilin (I or II) with increasing amounts of PIP2 (0- to 100-fold molar excess over profilin), and centrifuged them on a filter with a molecular weight cut-off of 30 000. The flow-through contains profilin not bound to PIP2 and is shown in Figure 1A. The result of scanning the gels is given in Figure 1B. With a 25-fold molar excess of PIP2 over profilin, nearly all profilin I is bound to PIP2. On the other hand, even with a 100-fold molar excess of PIP2, 60% of profilin II (relative to the sample without PIP2) is still present in the flow-through. This microfiltration assay clearly demonstrates the differences in affinity for PIP2 between the profilin isoforms. Figure 1.Microfiltration of profilin–PIP2 complexes. (A) Samples containing a constant amount of profilin (4 μM) and increasing concentrations of PIP2 were made. The flow-through was analysed by SDS–PAGE, of which only the 14–20 kDa region is shown. The molar excess of PIP2 is indicated. (B) Result of scanning the Coomassie-stained spots shown in (A). Boxes represent the percentage of non-bound profilin. White boxes are profilin I and black boxes profilin II. The amount of profilin found in the flow-through when no PIP2 is present was set to 100%. Download figure Download PowerPoint PIP2 binding induces a greater change in the structure of profilin I than in profilin II Local structure changes induced by phospholipids can be observed in near UV (250–320 nm) circular dichroism (CD) spectra, as has been shown for profilin I (Raghunathan et al., 1992) and the pleckstrin homology domain of spectrin (Hyvönen et al., 1995). The CD measurements were carried out for samples containing 15 μM profilin with different concentrations of PIP2 ranging from 0 to 300 μM (Figure 2). Profilin I and II alone have different spectra. This is probably because of the difference in their tyrosine, tryptophan and phenylalanine content and the environment of these residues. The ellipticity of profilin I in the presence of PIP2 shows a large decrease with only a 10-fold molar excess of PIP2 and only a slight further decrease when the PIP2 concentration is doubled. In contrast, the decrease in ellipticity of profilin II is smaller and nearly doubles when the PIP2 concentration is doubled. Figure 2.Circular dichroism spectra in the near UV of 15 μM profilin (thick line) I or II with 150 (....) and 300 μM (− − −) PIP2. The molar ellipticity per residue weight is shown as a function of the wavelength. Download figure Download PowerPoint Profilin I and II display different affinities for proline-rich peptides Whereas profilin I has a higher affinity for PIP2, profilin II binds more strongly to poly-L-proline (Lambrechts et al., 1995). Poly-L-proline as such is not present in living cells, though several proteins have proline-rich sequences. At the start of our investigations, no natural profilin ligands with stretches of proline residues were identified. However, a number of proteins were likely candidates for a profilin ligand. CAP contains six and five prolines in a row separated by a single glycine residue (Matviw et al., 1992). A repeat of four prolines preceded by a phenylalanine occurs three times in ActA, each time separated by 30 amino acids (Domann et al., 1992), and, more recently, a three times repeated sequence of five consecutive proline residues separated by a glycine was identified in VASP (Haffner et al., 1995). A shorter sequence containing only five proline residues is also found in the VASP sequence. As these three proteins have been implicated in actin dynamics or are associated with the actin cytoskeleton, we chemically synthesized peptides derived from human CAP and VASP and from Listeria ActA (Table I) and studied their interaction with both profilin isoforms. We incubated a constant amount of profilin (1 μM) with an increasing amount of peptide (0–55 μM) under physiological conditions and monitored binding by spectrofluorimetry (Figure 3). Figure 3.Binding of profilin I and II to different proline-rich peptides. The relative fluorescence change (rfc) is plotted against peptide concentration. The best fitting logarithmic curves are drawn as full lines. The profilin concentration is 1 μM in all samples, and the peptide concentration ranges from 0 to 55 μM. (A) Profilin I and peptides peptCAPwt (♦), peptVASPwt (□), peptVASPs (▾) and peptACTwt (▵). (B) Profilin II and peptides peptCAPwt (♦), peptVASPwt (□), peptVASPs (▾) and peptACTwt (▵). (C) Profilin II and CAP wild-type and mutant peptides peptCAPwt (♦), peptCAPΔ1 (⋄), peptCAPΔ2 (▪), and peptCAPΔ4 (▿). Download figure Download PowerPoint Table 1. Sequence of proline-rich peptides Peptide Sequence Protein Reference peptCAPwt Ac.SGPPPPPPGPPPPPVS.OH human CAP Matviw et al. (1992) peptVASPwt Ac.GPPPPPGPPPPPGPPPPPGL.OH human VASP Haffner et al. (1995) peptVASPs Ac.GGPPPPPGL.OH human VASP Haffner et al. (1995) peptACTwt Ac.FPPPPTD.OH Listeria ActA Domann et al. (1992) peptCAPΔ1 Ac.SGPPPPPGPPPPPVS.OH CAP mutant peptCAPΔ2 Ac.SGPPPPGPPPPPVS.OH CAP mutant peptCAPΔ4 Ac.SGPPGPPPPPVS.OH CAP mutant The ActA peptide, peptACTwt, which contains only four proline residues, induces no significant increase in fluorescence even at high peptide concentrations, indicating that it has a very low affinity for both of the profilin isoforms (Figure 3A and B). This is consistent with the finding of Zeile et al. (1996) that the ActA peptide does not bind to profilin at a concentration of 100 μM. The same is observed for peptVASPs, a short proline-rich sequence derived from VASP. The other peptides studied, peptCAPwt and peptVASPwt, show a change in the fluorescence intensity which means they do bind to profilin. We tried to fit a hyperbolic curve to the measured values but could not find a sufficiently good fit. Also, attempts to linearize the data using the Surewicz and Epand equation (1984) were unsuccessful. This suggests that the binding is heterogenous, as will be demonstrated further below. Nevertheless, the difference in the steepness of the curves in the low concentration range in Figure 3A and B suggests a difference in affinity of these peptides for each of the profilin isoforms. They appear to bind better to profilin II than to profilin I. As peptCAPwt and peptVASPwt (with 11 and 15 proline residues respectively) seem to have a comparable affinity and peptACTwt (with only four prolines) showed no binding, we synthesized increasingly shorter mutant CAP peptides (see Table I) to assay the minimal length required for binding (Figure 3C). PeptCAPΔ1 lacking one proline residue has almost the same affinity as peptCAPwt, while peptCAPΔ2 and peptCAPΔ4, lacking two and four prolines respectively, have a strongly reduced affinity compared with wild-type CAP peptide. Profilin II dimerizes upon binding of proline-rich peptides Modelling experiments suggest that one repeat of five proline residues is sufficient to fill the hypothetical poly-L-proline binding pocket (data not shown). We demonstrated above that two repeats are necessary for profilin binding. This prompted us to study whether dimers of profilin can be formed on the proline-rich peptides. Since profilin II appears to bind better to these peptides, we chose to study this isoform. Profilin II elutes at 47 min on a Superdex 200 gel filtration column (Figure 4a). With a 1.2-fold molar excess of peptVASPwt, profilin II shifts partly to an earlier position (Figure 4b). When more peptide is added in 12-fold molar excess over profilin II, a second shift can be observed (Figure 4c) relative to the profilin II peak. In the simplest scenario, the first shift is a 1:1 complex of profilin and peptVASPwt and the second, larger shift would then be a 2:1 complex consisting of a dimer of profilin bound to one peptide. We observed the same shift with profilin II and peptCAPwt (data not shown), but observed no shift when an 18-fold molar excess of the short peptVASPs is added to profilin II, indicating that no binding occurs (Figure 4d). Figure 4.Gel filtration of profilin peptide complexes. Only the relevant part of the chromatograms between 40 and 50 min is shown. (a) Profilin II alone at 33 μM, (b) profilin II (33 μM) and a 1.2-fold molar excess of peptVASPwt, (c) profilin II (30 μM) and a 12-fold molar excess of peptVASPwt, (d) profilin II (20 μM) and an 18-fold molar excess of peptVASPs. Detection was at 280 nm, a wavelength where the peptides do not absorb, but their presence was investigated using reversed phase HPLC (data not shown). The original absorption units to full scale are set at 0.1 for (a) and (b) and 0.2 for (c) and (d). Download figure Download PowerPoint A proline-rich peptide induces a conformational change in profilin II Local structural changes in proteins induced by peptide binding can also be observed in CD spectra in the region from 250 to 290 nm. We measured spectra for profilin I and II with three different concentrations of peptCAPwt (Figure 5), and the results suggest that aromatic amino acids are involved in peptide binding, consistent with results from mutational analysis of profilin I (Björkegren et al., 1993). Figure 5.Circular dichroism spectra in the near UV. Profilin I or profilin II at a concentration of 15 μM was measured in the absence (thick line) or presence of peptide peptCAPwt at a concentration of 20 (....), 50 (− − −) or 100 μM (thin line). The molar ellipticity per residue weight is shown as a function of the wavelength. The profilin I and II spectra are somewhat different from those in Figure 2, due to a different buffer composition and dilution effects. Download figure Download PowerPoint The first decrease in profilin II ellipticity is very large, while subsequent decreases at larger peptide concentrations are small. This suggests that a conformational change accompanies the binding of profilin II to peptCAPwt. On the other hand, for profilin I, the ellipticity decreases more gradually as the peptide concentration increases, consistent with a lower affinity for the peptide. The polyproline binding site of profilins is not accessible when PIP2 is present We also investigated whether PIP2 and poly-L-proline compete for profilin binding. In the control experiment, profilin I or II were bound to poly-L-proline and, after washing, were eluted with 8 M urea. A small fraction of profilin I did not bind or bound very weakly to the poly-L-proline column. The majority of profilin I is in the early eluting 8 M urea fractions (Figure 6A, lanes 5 and 6) while the majority of profilin II elutes later (lanes 7–11). This result is consistent with our previously published data where a similar experiment was carried out at a different pH (Lambrechts et al., 1995) and again points to a difference in affinity for poly-L-proline of these two isoforms. Figure 6.Competitive interaction between PIP2 and poly-L-proline for the profilin isoforms. (A) Profilin I (white boxes) and II (black boxes) are loaded onto a poly-L-proline column and eluted with 8 M urea. Fractions 1 and 2, flow-through; 3 and 4, wash; 5–12, 8 M urea elution. (B) Profilin I (white boxes) and II (black boxes) are bound to a poly-L-proline column and eluted with PIP2. The PIP2 gradient is indicated by full lines. Fraction 1, flow-through; 2, wash; 3, 50 μM PIP2; 4, 100 μM PIP2; 5, 200 μM PIP2; 6, 300 μM PIP2; 7, 400 μM PIP2; 8, wash; 9–12, 8 M urea. (C) Samples containing profilin I (white boxes) or II (black boxes), pre-incubated with a 25-fold molar excess of PIP2, are loaded on a poly-L-proline column and eluted with 8 M urea. Fraction 1 and 2, flow-through; 3–10, buffer wash; 11–16, 8 M urea elution. Download figure Download PowerPoint We next loaded the poly-L-proline affinity column with each profilin isoform and, after washing, eluted them with increasing concentrations of PIP2 (Figure 6B). We observed that PIP2 was capable of eluting profilin I from the column although a significant amount still remained bound to poly-L-proline. By contrast, PIP2 was not able to elute profilin II. In the reverse experiment, the profilin isoforms were incubated with a 25-fold molar excess of PIP2 micelles prior to loading them onto the column. A large fraction of profilin I is found in the flow-through and the wash (combined, ∼53%), though still ∼47% is eluted with 8 M urea (Figure 6C). In the profilin II sample, only a small amount of profilin is found in the flow-through (17%). No profilin II is found in the wash, and the remaining 83% can only be eluted with 8 M urea. These results show that profilin I has a higher affinity for PIP2 than for poly-L-proline. Identification of VASP as a ligand for profilin II but not for profilin I Reinhard and colleagues (1995) identified VASP as a ligand for profilins and, since our experiments with the VASP-derived peptide suggest that the profilin isoforms have a different affinity for this ligand, we assayed the binding of VASP to each isoform using profilin I or II affinity columns. We first show that the monoclonal antibodies against human VASP also recognize bovine VASP (Figure 7A). Figure 7.Binding of VASP to profilin. (A) Immunoprecipitation of bovine VASP from brain extracts with monoclonal antibodies against human VASP. (B) Profilin affinity chromatography and Western blotting using anti-VASP monoclonal antibody. Profilin I or II Sepharose columns are loaded with bovine brain extracts and, after washing, eluted with increasing concentrations of peptVASPwt. The eluted fractions are precipitated in 10% trichloroacetic acid and, after gel electrophoresis and Western blotting, tested for the presence of VASP using monoclonal antibodies. Download figure Download PowerPoint We loaded an equal amount of bovine brain extract on the columns and, after washing, eluted the remaining proteins with a step gradient of the proline-rich peptide derived from VASP (peptVASPwt) as indicated in Figure 7B. We tested the eluted fractions for the presence of VASP on Western blot using a monoclonal antibody (Figure 7B). Interestingly the capacity of the profilin isoforms to retain VASP appears different. While only a barely noticeable amount of VASP is eluted from the profilin I column, a much larger amount of VASP bound to the profilin II column. The elution starts with 50 μM peptVASPwt but the majority of the protein is eluted with 500 μM peptVASPwt. This result confirms the other in vitro experiments with purified proteins and peptides and indicates that profilin II rather than profilin I is a possible ligand for VASP. Discussion In this study, we demonstrate that the profilin isoforms have complementary affinities for two ligands. Profilin I interacts strongly with PIP2 and more weakly with poly-L-proline sequences. On the other hand, profilin II has a higher affinity for proline-containing peptides and binds less tightly to PIP2. Our data on binding of profilin I to proline-rich peptides are in agreement with previous results of Perelroizen and co-workers (1994). These authors postulated the optimal length of the poly-L-proline sequence to be 15–20 residues. However, we find that there is little difference in affinity between peptCAPwt and peptVASPwt containing 11 and 15 proline residues, and a slight decrease in affinity of a mutant CAP peptide lacking one proline (peptCAPΔ1). Shorter peptides, such as those that occur in the Listeria surface protein ActA and the shorter repeats in VASP, have very low affinity for both of the profilin isoforms. This result suggests that the role of profilin in Listeria movement is not by direct interaction with ActA, consistent with recent results showing that VASP may be the host factor that links profilin to ActA (Chakraborty et al., 1995). The precise cellular function of VASP, a focal adhesion protein, is not known but its interaction with profilin, which is known from in vitro experiments, suggests that it may be an anchor for profilin-mediated actin polymerization (Pantaloni and Carlier, 1993; Reinhard et al., 1995). This is exemplified by the fact that after Listeria infection, VASP is recruited to the bacterial surface and binds ActA, prior to actin polymerization (Chakraborty et al., 1995), and that depletion of profilin from extracts used for in vitro motility assays slows down Listeria bacteria (Theriot et al., 1994). So far, a direct binding of CAP to profilin has not been demonstrated, although a functional link between the two proteins exists in yeast (Vojtek et al., 1991). Overexpression of profilin is able to rescue morphological defects associated with deletion of the CAP C-terminal domain. Interestingly the part of CAP still present in this mutant contains the proline-rich sequence, used in the present study. This suggests that the poly-L-proline site may be cryptic most of the time and only made accessible after an, as yet unknown, type of regulation. Cryptic sites have been identified in other cytoskeletal proteins (Menkel et al., 1994; Turunen et al., 1994; Gilmore and Burridge, 1995). We also found that profilin II forms dimers upon binding to proline-rich peptides. If this also occurs in vivo it gives the cell a more efficient machinery to direct localized actin assembly. For each tetrameric VASP molecule (Haffner et al., 1995) eight profilin molecules would be recruited at focal adhesions (or other sites) where VASP is active. In view of the polymerization-promoting activity of profilin (Pantaloni and Carlier, 1993), this concentrating effect could have a dramatic effect on the actin dynamics at these subcellular localizations. This dimerization may also explain the slightly sigmoidal behaviour of the fluorescence data, from which it appears that profilin II has a higher affinity for proline-rich peptides than has profilin I. This is evidenced further by other experiments in the present study. An excess of peptCAPwt only induces a small change in the near UV CD spectrum of profilin I, while similar concentrations result in a large change in the CD spectrum of profilin II. In addition, elution of profilin I from poly-L-proline affinity columns requires less urea than does profilin II (control experiment in Figure 5 and Lambrechts et al., 1995). Finally, in brain extracts, we could recover VASP only from profilin II–Sepharose columns (see also below). Although the profilins from eukaryotic organisms display this difference in affinity, in Acanthamoeba both profilin isoforms appear to have similar dissociation constants for poly-L-proline (Kaiser and Pollard, 1996). We also show that, in contrast to poly-L-proline binding, profilin I has a higher affinity for PIP2 than does profilin II. Using a 50-fold molar excess of PIP2 over profilin, all of isoform I and only ∼24% of isoform II was bound to PIP2 vesicles. The CD spectra after PIP2 binding show a large decrease for profilin I and only a small decrease for profilin II. This may indicate that profilin II binding to PIP2 is not accompanied by structural changes or, more probably, that only a small amount of profilin II is bound to PIP2. The comp" @default.
• W2022122132 created "2016-06-24" @default.
• W2022122132 creator A5003890622 @default.
• W2022122132 creator A5022472003 @default.
• W2022122132 creator A5041982680 @default.
• W2022122132 creator A5055228194 @default.
• W2022122132 creator A5067759425 @default.
• W2022122132 creator A5075879769 @default.
• W2022122132 date "1997-02-01" @default.
• W2022122132 modified "2023-09-27" @default.
• W2022122132 title "The mammalian profilin isoforms display complementary affinities for PIP2 and proline-rich sequences" @default.
• W2022122132 cites W1482377527 @default.
• W2022122132 cites W155617828 @default.
• W2022122132 cites W1566160921 @default.
• W2022122132 cites W1574312817 @default.
• W2022122132 cites W1741161055 @default.
• W2022122132 cites W1921408963 @default.
• W2022122132 cites W1928927063 @default.
• W2022122132 cites W1966791529 @default.
• W2022122132 cites W1969255709 @default.
• W2022122132 cites W1970195231 @default.
• W2022122132 cites W1972467203 @default.
• W2022122132 cites W1975541739 @default.
• W2022122132 cites W1976121619 @default.
• W2022122132 cites W2022129165 @default.
• W2022122132 cites W2036267702 @default.
• W2022122132 cites W2044302013 @default.
• W2022122132 cites W2045590135 @default.
• W2022122132 cites W2047831969 @default.
• W2022122132 cites W2070947408 @default.
• W2022122132 cites W2073125687 @default.
• W2022122132 cites W2076300921 @default.
• W2022122132 cites W2079420220 @default.
• W2022122132 cites W2082705660 @default.
• W2022122132 cites W2087735445 @default.
• W2022122132 cites W2090350631 @default.
• W2022122132 cites W2090430644 @default.
• W2022122132 cites W2112811719 @default.
• W2022122132 cites W2117822148 @default.
• W2022122132 cites W2130592994 @default.
• W2022122132 cites W2149624337 @default.
• W2022122132 cites W2167948821 @default.
• W2022122132 doi "https://doi.org/10.1093/emboj/16.3.484" @default.
• W2022122132 hasPubMedCentralId "https://www.ncbi.nlm.nih.gov/pmc/articles/1169652" @default.
• W2022122132 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/9034331" @default.
• W2022122132 hasPublicationYear "1997" @default.
• W2022122132 type Work @default.
• W2022122132 sameAs 2022122132 @default.
• W2022122132 citedByCount "136" @default.
• W2022122132 countsByYear W20221221322012 @default.
• W2022122132 countsByYear W20221221322013 @default.
• W2022122132 countsByYear W20221221322014 @default.
• W2022122132 countsByYear W20221221322015 @default.
• W2022122132 countsByYear W20221221322016 @default.
• W2022122132 countsByYear W20221221322017 @default.
• W2022122132 countsByYear W20221221322018 @default.
• W2022122132 countsByYear W20221221322019 @default.
• W2022122132 countsByYear W20221221322020 @default.
• W2022122132 countsByYear W20221221322021 @default.
• W2022122132 countsByYear W20221221322023 @default.
• W2022122132 crossrefType "journal-article" @default.
• W2022122132 hasAuthorship W2022122132A5003890622 @default.
• W2022122132 hasAuthorship W2022122132A5022472003 @default.
• W2022122132 hasAuthorship W2022122132A5041982680 @default.
• W2022122132 hasAuthorship W2022122132A5055228194 @default.
• W2022122132 hasAuthorship W2022122132A5067759425 @default.
• W2022122132 hasAuthorship W2022122132A5075879769 @default.
• W2022122132 hasBestOaLocation W20221221321 @default.
• W2022122132 hasConcept C104317684 @default.
• W2022122132 hasConcept C142669718 @default.
• W2022122132 hasConcept C1491633281 @default.
• W2022122132 hasConcept C203253989 @default.
• W2022122132 hasConcept C2779815552 @default.
• W2022122132 hasConcept C2780283098 @default.
• W2022122132 hasConcept C2993400109 @default.
• W2022122132 hasConcept C515207424 @default.
• W2022122132 hasConcept C53345823 @default.
• W2022122132 hasConcept C54355233 @default.
• W2022122132 hasConcept C55493867 @default.
• W2022122132 hasConcept C78458016 @default.
• W2022122132 hasConcept C86803240 @default.
• W2022122132 hasConceptScore W2022122132C104317684 @default.
• W2022122132 hasConceptScore W2022122132C142669718 @default.
• W2022122132 hasConceptScore W2022122132C1491633281 @default.
• W2022122132 hasConceptScore W2022122132C203253989 @default.
• W2022122132 hasConceptScore W2022122132C2779815552 @default.
• W2022122132 hasConceptScore W2022122132C2780283098 @default.
• W2022122132 hasConceptScore W2022122132C2993400109 @default.
• W2022122132 hasConceptScore W2022122132C515207424 @default.
• W2022122132 hasConceptScore W2022122132C53345823 @default.
• W2022122132 hasConceptScore W2022122132C54355233 @default.
• W2022122132 hasConceptScore W2022122132C55493867 @default.
• W2022122132 hasConceptScore W2022122132C78458016 @default.
• W2022122132 hasConceptScore W2022122132C86803240 @default.
• W2022122132 hasIssue "3" @default.
• W2022122132 hasLocation W20221221321 @default.
• W2022122132 hasLocation W20221221322 @default.
• W2022122132 hasLocation W20221221323 @default.

• next